def runTest(self):
resolver = Resolver()
# workspace = resolver.workspace_from_url(assets.url_of('SBB0000F29300010000/mets_one_file.xml'), directory=WORKSPACE_DIR)
workspace = resolver.workspace_from_url(
assets.url_of('kant_aufklaerung_1784-binarized/mets.xml'),
directory=WORKSPACE_DIR)
TesserocrSegmentRegion(workspace,
input_file_grp="OCR-D-IMG",
output_file_grp="OCR-D-SEG-BLOCK").process()
TesserocrSegmentLine(workspace,
input_file_grp="OCR-D-SEG-BLOCK",
output_file_grp="OCR-D-SEG-LINE").process()
TesserocrSegmentWord(workspace,
input_file_grp="OCR-D-SEG-LINE",
output_file_grp="OCR-D-SEG-WORD").process()
workspace.save_mets()
def runTest(self):
resolver = Resolver(cache_enabled=True)
# workspace = resolver.workspace_from_url(assets.url_of('SBB0000F29300010000/mets_one_file.xml'), directory=WORKSPACE_DIR)
workspace = resolver.workspace_from_url(assets.url_of(
'kant_aufklaerung_1784-page-block-line-word/mets.xml'),
directory=WORKSPACE_DIR)
TesserocrSegmentRegion(workspace,
input_file_grp="OCR-D-IMG",
output_file_grp="OCR-D-SEG-BLOCK").process()
workspace.save_mets()
TesserocrSegmentLine(workspace,
input_file_grp="OCR-D-SEG-BLOCK",
output_file_grp="OCR-D-SEG-LINE").process()
workspace.save_mets()
TesserocrRecognize(workspace,
input_file_grp="OCR-D-SEG-LINE",
output_file_grp="OCR-D-OCR-TESS",
parameter={
'textequiv_level': 'word'
}).process()
workspace.save_mets()
def runTest(self):
resolver = Resolver()
# workspace = resolver.workspace_from_url(assets.url_of('SBB0000F29300010000/mets_one_file.xml'), directory=WORKSPACE_DIR)
workspace = resolver.workspace_from_url(assets.url_of(
'kant_aufklaerung_1784-page-block-line-word/mets.xml'),
directory=WORKSPACE_DIR)
TesserocrSegmentRegion(workspace,
input_file_grp="OCR-D-IMG",
output_file_grp="OCR-D-SEG-BLOCK").process()
workspace.save_mets()
TesserocrSegmentLine(workspace,
input_file_grp="OCR-D-SEG-BLOCK",
output_file_grp="OCR-D-SEG-LINE").process()
workspace.save_mets()
TesserocrRecognize(
workspace,
input_file_grp="OCR-D-SEG-LINE",
output_file_grp="OCR-D-OCR-TESS",
parameter={
'textequiv_level': 'line'
} # add dep tesseract-ocr-script-frak: , 'model': 'Fraktur'
).process()
workspace.save_mets()
TesserocrSegmentWord(workspace,
input_file_grp="OCR-D-SEG-LINE",
output_file_grp="OCR-D-SEG-WORD").process()
workspace.save_mets()
TesserocrRecognize(
workspace,
input_file_grp="OCR-D-SEG-WORD",
output_file_grp="OCR-D-OCR-TESS-W2C",
parameter={
'textequiv_level': 'glyph'
} # add dep tesseract-ocr-script-frak: , 'model': 'Fraktur'}
).process()
workspace.save_mets()
import os
import shutil
from test.base import TestCase, main, assets
from ocrd.resolver import Resolver
from ocrd_tesserocr import TesserocrSegmentRegion
METS_HEROLD_SMALL = assets.url_of('SBB0000F29300010000/data/mets_one_file.xml')
WORKSPACE_DIR = '/tmp/pyocrd-test-segment-region-tesserocr'
class TestTesserocrSegmentRegionTesseract(TestCase):
def setUp(self):
if os.path.exists(WORKSPACE_DIR):
shutil.rmtree(WORKSPACE_DIR)
os.makedirs(WORKSPACE_DIR)
def runTest(self):
resolver = Resolver()
workspace = resolver.workspace_from_url(METS_HEROLD_SMALL,
dst_dir=WORKSPACE_DIR)
TesserocrSegmentRegion(workspace,
input_file_grp="OCR-D-IMG",
output_file_grp="OCR-D-SEG-BLOCK").process()
workspace.save_mets()
if __name__ == '__main__':
main()
# pylint: disable=import-error
import os
import shutil
from tempfile import TemporaryDirectory
from test.base import TestCase, assets, main
from ocrd.resolver import Resolver
from ocrd_ocropy.segment import OcropySegment
PARAM_JSON = assets.url_of('param-segment.json')
WORKSPACE_DIR = '/tmp/ocrd-ocropy-segment-test'
class TestOcropySegment(TestCase):
def setUp(self):
if os.path.exists(WORKSPACE_DIR):
shutil.rmtree(WORKSPACE_DIR)
os.makedirs(WORKSPACE_DIR)
def test_run1(self):
resolver = Resolver()
with TemporaryDirectory() as tempdir:
workspace = resolver.workspace_from_url(assets.path_to(
'kant_aufklaerung_1784-binarized/data/mets.xml'),
dst_dir=tempdir)
proc = OcropySegment(
workspace,
input_file_grp="OCR-D-IMG-BIN",
def setUp(self):
self.mets = OcrdMets(
filename=assets.url_of('SBB0000F29300010000/mets.xml'))
def setUp(self):
self.resolver = Resolver()
self.workspace = self.resolver.workspace_from_url(assets.url_of('SBB0000F29300010000/mets.xml'))
import os
import shutil
from test.base import TestCase, main, assets, skip
from ocrd.resolver import Resolver
from ocrd_tesserocr.segment_word import TesserocrSegmentWord
from ocrd_tesserocr.segment_line import TesserocrSegmentLine
from ocrd_tesserocr.segment_region import TesserocrSegmentRegion
from ocrd_tesserocr.recognize import TesserocrRecognize
#METS_HEROLD_SMALL = assets.url_of('SBB0000F29300010000/data/mets_one_file.xml')
METS_HEROLD_SMALL = assets.url_of('kant_aufklaerung_1784/data/mets.xml')
WORKSPACE_DIR = '/tmp/pyocrd-test-recognizer'
class TestTesserocrRecognize(TestCase):
def setUp(self):
if os.path.exists(WORKSPACE_DIR):
shutil.rmtree(WORKSPACE_DIR)
os.makedirs(WORKSPACE_DIR)
#skip("Takes too long")
def runTest(self):
resolver = Resolver()
workspace = resolver.workspace_from_url(METS_HEROLD_SMALL,
dst_dir=WORKSPACE_DIR)
TesserocrSegmentRegion(workspace,
input_file_grp="OCR-D-IMG",
output_file_grp="OCR-D-SEG-BLOCK").process()
def runTest(self):
report = WorkspaceValidator.validate_url(
self.resolver,
assets.url_of('SBB0000F29300010000/mets_one_file.xml'))
print(report.to_xml())
def runTest(self):
resolver = Resolver(cache_enabled=True)
workspace = resolver.workspace_from_url(
assets.url_of('SBB0000F29300010000/mets_one_file.xml'))
TesserocrSegmentRegion(workspace).process()
workspace.save_mets()
import os
from shutil import copytree, rmtree
from ocrd.model import OcrdExif
from ocrd.resolver import Resolver
from test.base import TestCase, assets, main
TMP_FOLDER = '/tmp/test-pyocrd-resolver'
METS_HEROLD = assets.url_of('SBB0000F29300010000/mets.xml')
FOLDER_KANT = assets.url_of('kant_aufklaerung_1784')[len('file://'):]
TEST_ZIP = assets.url_of('test.ocrd.zip')[len('file://'):]
class TestResolver(TestCase):
def setUp(self):
self.resolver = Resolver(cache_enabled=True)
self.folder = os.path.join(TMP_FOLDER, 'kant_aufklaerung_1784')
if os.path.exists(TMP_FOLDER):
rmtree(TMP_FOLDER)
os.makedirs(TMP_FOLDER)
copytree(FOLDER_KANT, self.folder)
def test_unpack_workspace(self):
workspace = self.resolver.unpack_workspace_from_filename(TEST_ZIP)
files = workspace.mets.find_files(mimetype='image/tiff')
self.assertEqual(len(files), 2, '2 TIF')
for f in files:
workspace.download_file(f)
print([OcrdExif.from_filename(f.local_filename).to_xml() for f in files])
def test_workspace_from_folder(self):
import os
from shutil import copytree, rmtree
from test.base import TestCase, assets, main
from ocrd.model import OcrdExif
from ocrd.resolver import Resolver
TMP_FOLDER = '/tmp/test-pyocrd-resolver'
METS_HEROLD = assets.url_of('SBB0000F29300010000/mets.xml')
FOLDER_KANT = assets.path_to('kant_aufklaerung_1784')
TEST_ZIP = assets.path_to('test.ocrd.zip')
class TestResolver(TestCase):
def setUp(self):
self.resolver = Resolver()
self.folder = os.path.join(TMP_FOLDER, 'kant_aufklaerung_1784')
if os.path.exists(TMP_FOLDER):
rmtree(TMP_FOLDER)
os.makedirs(TMP_FOLDER)
copytree(FOLDER_KANT, self.folder)
def test_workspace_from_url(self):
workspace = self.resolver.workspace_from_url(METS_HEROLD)
# print(METS_HEROLD)
# print(workspace.mets)
input_files = workspace.mets.find_files(fileGrp='OCR-D-IMG')
# print [str(f) for f in input_files]
image_file = input_files[0]
# print(image_file)
f = workspace.download_file(image_file)
